Pick one: Works, Fast, or Right

Works, fast, right. On any given task you can usually optimise for one, sometimes two, almost never all three at once. The engineers who get into trouble are the ones who pretend the trade-off isn't there.

The three pulls

• Works: it does the thing, in front of the user, today. Correctness of outcome, right now.

• Fast: it's delivered quickly — shipped this week, not next quarter.

• Right: it's correct, clean, and maintainable; it won't bite you in six months.

You can't max all three at once

Fast and Works gets you a demo and a mountain of tech debt. Right and Works gets you something solid that arrives too late to matter. Fast and Right that also works first time is a unicorn — and usually means the problem was easy and you're flattering yourself. Pick the corner you're aiming for honestly.

Choose deliberately, not by accident

The sin isn't trading off — trading off is the job. The sin is not knowing which one you chose. A throwaway prototype should aim squarely at Works-and-Fast, and everyone in the room should know it's disposable. A payments system aims at Right and accepts it'll take longer. Disaster is when you build the prototype with no awareness it's a prototype, and it quietly becomes the thing running in production for five years.

Make it work, then right, then fast

Kent Beck's old order still holds, just stretched over time rather than crammed into one moment: get it working, then get it right, then — only if you measured and it matters — get it fast. You can't have all three today. You can have all three eventually, in that sequence, if you're deliberate about it.